This proposal outlines the development of a general technique to identify human monoclonal antibodies (mAbs). Murine mAbs are being used in many clinical applications such as cancer, septic shock, and transplantation. The use of human mAbs would obviate the human anti-mouse mAb response which often blunts the clinical usefulness of mAb immunotherapy. However, for most indications human mAbs are very difficult or impossible to obtain against human antigens. The use of hyperimmunized SCID-hu mice as a source of human B-cells, in combination with bacteriophage lambda immunoexpression cloning techniques, should yield human mAbs against selected human antigens. Human mAbs made by the proposed method would not have the same antigenic restrictions as previous methods and thus could be of great practical utility.